Strategy ONE

Create an Agent Using a Template

Starting in Strategy One (September 2025), you can create a retail analytics agent by selecting a pre-built template. The template's workflow guides you through:

  • Connecting a dataset

  • Mapping their data objects to the agent’s required schema

  • Adding aliases for non-standard naming

  • Removing unnecessary objects

The workflow supports extensibility, allowing you to edit the agent’s AI dataset collection after you create it, including adding new objects.

Out-of-the-box reliability is guaranteed for the validated template configuration, not for objects added to it.

Prerequisites

  • Ensure that you have the correct privileges, as outlined in Prerequisites for Running and Creating Agents.

  • Your project must contain at least one AI-enabled dataset.

    An agent's data is provided by an AI dataset collection. When you use a template, you create a new collection. Enabling a dataset for AI enriches it with automatically-generated descriptions of the cube itself and each column, providing the context needed for your agents. Use Workstation to enable datasets.

Create an Agent Using the Retail Agent Template

  1. Click Create New and select Agent.
  2. If you have access to multiple projects, from the Create Agent In drop-down list, select the project to create the agent in.
  3. In the New Agent window, click Select Template.
  4. Click Retail Agent Template, then click Create.
  5. Select a dataset and click Create.
    • The template works with a single dataset.

    Your dataset objects are automatically mapped to the template's objects. For example, the template contains Payment Method which is automatically mapped to the dataset object Payment Type DESC.

    • Dataset objects that are mapped to the template are bolded in the My Objects panel.

    • A mapped dataset object uses the template object's description. Object descriptions are pre-defined and cannot be modified, which ensures clarity and consistency.

  6. If an object has been incorrectly mapped, click the X in the attribute in the Replace With column.
  7. To map an object, drag it from the My Objects panel to the Replace With column of the correct template object form. For example, Date was not automatically mapped to any dataset object. Drag Transaction Date from the My Objects panel to the Date in the Template Object Forms.

    • You can map objects that have different data types. For example, you can replace Customer Segment, which is text, with a numeric Customer Group ID.

    • You cannot use a metric to replace an attribute form, or an attribute form to replace a metric.

  8. If you do not want to include a template object in the agent, leave the Replace With column blank. If the object was automatically mapped, remove the dataset object by clicking the X in the attribute in the Replace With column.

    For example, Product ID is mapped, but Product Name is not available in the dataset and is not mapped to an object. The template object is blank.

  9. You can add and remove aliases for non-standard naming.
    • To add an alias, click in the Aliases column and type the alias.
    • To delete an automatic alias, click in the Aliases column and then click the X for the alias.
  10. Click Launch in Agent.
  11. If key columns are missing, a warning message displays. Missing columns can affect data accuracy. Click Continue to accept or Cancel to continue mapping objects.
  12. Save your agent:

    1. Type a Name for your agent.

    2. Type an optional Description.

    3. Select the folder to save it in.

    4. Click Save.

    The new agent displays, ready for you to ask questions.

  13. To customize the agent, click the Edit icon . Customize the agent beginning at this step.